github-code-review

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests user-generated GitHub PR content (e.g., "gh pr view 123 --json files..." and "PR_DIFF=$(gh pr diff 123)" in the Usage Examples and GitHub Actions sections of SKILL.md) and passes that untrusted content into review agents (npx ruv-swarm) whose outputs are used to post reviews/approve/request-changes, so third-party PR text/diffs can materially influence agent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes "npx ruv-swarm" at runtime which will fetch and execute the remote npm package (e.g. https://registry.npmjs.org/ruv-swarm), meaning externally-hosted code can run and control the agent's behavior.